J, Agric. Food Chem. 1988, 36, 293-295
study on triglycerides could lead to more complete information and more valuable results for avocado maturity determination. Such research work is worth undertaking. ACKNOWLEDGMENT
Technical assistance in collecting fruits from the Agronomic Station of Corsica Island has been provided by M. Vogel. We express our gratitude to J. R. Llinas for his help in statistical analyses. LITERATURE CITED Afnor Recueil des Normes FranGaises, Corps Gras, Graines OlLagineuses, Produits DBriuLs, 2nd ed.; Afnor: Paris, 1981. Ahmed, E. M. Barmore, C. R. Tropical and Subtropical Fruits; Nagy, S., Shaw, P. E., Eds; Avi: West Port, CT, 1980; p 121. Foucart, T. Analyse Factorielle; Masson: Paris, 1982. Gaydou, E. M.; Rasoarahona, J.; Bianchini, J. P. J. Sci. Food Agric. 1983,34, 1130. Gaydou, E. M.; Iatrides, M. C.; Ramananarivo, R.; Artaud, J.; Peiffer, G. J. Agric. Food Chem. 1984, 32, 651. Gaydou, E. M.; Ralambofetra, E.; Rakotovao, L.; Llinas, J. R. Analusis 1985, 13, 379.
293
Lebart, L.; Morineau, A.; FBnelon, J. P. Traitements de DonnLes Statistiques. MLthodes et Programmes; Dunod Paris, 1982. Lewis, C. E. J. Sci. Food Agric. 1978, 29, 857. Lewis, C. E.; Morris, R.; O'Brien, K. J. Sci. Food Agric. 1978, 29, 943. Lozano, Y.; Duverneuil, G.; Yamasaki, K. Fruits 1982,37, 545. Lozano, Y.; Ratovohery, J.; Gaydou, E. M. Rev. Fr. Corps Gras 1985, IO, 377. Lozano, Y. F.; Ratovohery, J. V.; Gaydou, E. M. Fruits 1987,42, 305.
Mazliak, P. The Biochemistry of Fruits a n d Their Products; Hulme, A. C., Ed.; Academic: London, 1970; p 209. Metcalfe, L. D.; Wang, C. N. J. Chromatogr. Sci. 1981,19, 530. Peyronnel, D.; Artaud, J.; Iatrides, M. C.; Rancurel, P.; Chevalier, J. L. Lipids 1984, 19, 643. Romeder, J. M. Mgthodes et Programmes d'AnaEyse Discriminante; Bordas: Paris, 1973. Received for review November 25,1986. Accepted July 7, 1987. This work was supported in part by the Fonds #Aide et de Cooperation (France) and by a research grant from the CIES for J.V.R.
Selenium Content of Vegetables, Fruits, and Cereals in Galicia (Northwest Spain) M. Mejuto-Mart:, M. Bollah-Rodriguez, C. Herrero-Latorre, and F. Bermejo-Mardnez*
The selenium content of 60 samples of vegetable foods has been determined fluorimetrically. After wet destruction of organic matter, the complex 4,5-benzopiazselenole was formed by reaction between Se(IV) and 2,3-diaminonaphthalene and extracted with cyclohexane. The results suggest that the population of Galicia (NW Spain) consumes an average 11.20 pg of selenium/person pea day in food of vegetable origin.
The nutritional value of selenium has been studied in experimental animals for ever 20 years (Schwarz and Foltz, 1957), and there has been growing interest in its role as an essential trace element for animals (NRC Agricultural Board, 1971) and for man (Levander, 1975). It has been found, for example, that selenium-supplemented diets effectively prevent the appearance of certain nutritional disorders in battery chickens and rats (Hartley, 1959). According to Morris and Levander (1970), the selenium content of human food it is varied. Vegetables and fruit are generally poor in selenium (0.01 pg/g), though mushrooms (Pipponen, 1984), garlic (Olson and Palmer, 1984), and horseradish (Levander, 1976) can contain quite large amounts. The selenium content of plants is in any case strongly influenced by the quantity of biologically available selenium in the soil in which they grow (Kubota et al., 1967), and hence by their geographical origin. For example, Mondragon and Jaffe (1976) have found that various food products contain higher selenium levels when acquired in the marketplace in Caracas than when bought in the United States. Departamento de Quimica Analitica, Facultad de Quimica, Universidad de Santiago de Compostela, Santiago de Compostela, Spain. 0021-8561/88/1436-0293$01.50/0
With respect to their intrinsic capacity for accumulation of Se, plants may be divided in three groups (Rosenfield apd Beath, 1964): group I, plants that may contain up to 10OOO pg/g; group 11, plants rarely containing more than a few hundred micrograms/gram; group 111, plants rarely containing more than about 30 pg/g, among them fruit and cereals. MATERIALS AND METHODS
Apparatus. Ordinary laboratory glassware was employed. In order to eliminate fluorescence due to detergents and samples, all glassware was washed after each use with tap water, 1:l nitric acid, and distilled water. Fluorimetric determinations were carried out on a Farrand Model A4 fluorimeter (Farrand Optical Co. Inc.) equipped with a mercury vapor lamp and 1-cm light path cells. Reagents. Selenium Standard Solution. A vial of Merck Titrisol selenium standard solution containing 1.OOO f 0.002 g of Se is made up to 1L, with deionized distilled water. Working solutions are derived from this standard by successive dilutions with 0.1 M HC1. 2,3-Diaminonaphthalene (DAN) Solution. Solutions (0.1%, (w/v) of DAN in 0.1 M HC1 are made up daily and extracted before use with 25 mL of cyclohexane/ 100 mL of DAN to eliminate fluorescence. 0 1988 American Chemical Society
294
J. Agric. Food Chem., Vol. 36, No. 2, 1988 04 0
Mejuto-Marti et al.
r
Table I. Reproducibility of the Method std dev of 10 YO sample determinations, ba/mL error fruit (Punica granatum) 1.35 x 10-3 1.77 mushroom (Lactarius piperatus) 1.49 X 2.19 cereal (wheat flour) 4.65 x 10-4 25.43
I
z u w
03
/
I
I
0,OO
/-
I 0,02 )lg
0,04
0,06
0,OS
0,l
Se(IV)/ml
Figure 1. Accuracy of the method used for selenium determination applicated for Actinidia chinensis, as determined by the method of additions.
EDTA-Hydroxylamine Hydrochloride Solution. EDTANaz.2Hz0 (0.3 g) and 25 g of hydroxylamine hydrochloride are dissolved in distilled water, and the solution is made up to 1 L. Samples. Sixty samples of food of vegetable origin were obtained from food shops or from growers (36 of fruit, 6 of nuts, 7 of cereals or flour, 8 of mushrooms, 2 of bread, 2 of cocoa). The samples were dried at 100-105 OC for 2-4 h depending on their humidity and then ground manually to a fine powder in a mortar. Two 1-g subsamples of each were taken for analysis. Procedure. After digestion with a 4:l mixture of nitric and perchloric acids and a 1:l mixture of nitric and sulfuric acids, by the method of Analytical Methods Committee, Royal Society of Chemistry (1979), samples were diluted to 50 mL with distilled water and boiled gently for at least 5 min with 5 mL of 37% HC1. After cooling, 5 mL of 50% formic acid and 10 mL of the EDTA-hydroxylamine hydrochloride solution were added to eliminate interferences and bring acidity within the range pH 1.5-2.0 with about 35 mL of a 1:l mixture of ammonia and water. 43Benzopiazselenolewas formed by adding 5 mL of the 0.1% DAN solution and warming in a water bath at 50 "C for 30 min. The mixture was allowed to cool, transferred to decanting funnels, extracted with 10 mL of cyclohexane, and washed with 25 mL of 0.1 M HC1, and the fluorescence of the complex was measured. Calibration curves must be prepared for each series of measurements employing the same procedure, but with organic matter being destroyed with just 1mL of perchloric acid. Reproducibility and Accuracy. The reproducibility of the Se measurements in the present study was determined by using the procedure described above to measure the Se content of each of a series of random food samples 10 times. The accuracy of the method was estimated by addition: for each of a series of random food samples, the concentration of Se was determined both without addition of Se standard and after various quantities of standard has been added. RESULTS AND DISCUSSION
The use of fluorimetry for the determination of selenium has been recommended in recent years by a variety of authorities including the Analytical Methods Committee of the Royal Society of Chemistry (1979) and the IUPAC (1984). Though the need for temperature control, the slow digestion process, and the extraction of the complex from samples and standards by shaking for 1 min with cyclohexane make this technique lengthier than some other methods referred to by Ihnat (1976), 2,3-DAN has proved to be a sensitive linear reagent for the determination of selenium in food when interferences from Al, Fe, Cu, etc.,
Table 11. Accuracy of the Method Se4+added, sample PPm Actinidia chinensis 0.01 0.02 0.03 0.03 0.05 Lactarius piperatus 0.10 0.20 0.30 0.40 0.50 wheat flour 0.10 0.20 0.30 0.40 0.50
Se4+found, PPm 0.085 0.095 0.104 0.115 0.128 0.132 0.500 0.600 0.720 0.820 0.925 1.000 0.020 0.120 0.235 0.320 0.425 0.525
recovery, YO
100.00 99.05 100.00 102.40 97.78 100.00 102.86 102.50 102.78 100.00 100.00 102.27 100.00 101.19 100.96
Table 111. Selenium Content of Vegetable Food in Galicia samde no. of samrdes ranee" mean" fruit 36 0.01-0.08 0.015 mushroom* 8 0.20-1.80 0.690 7 0.015-0.09 0.034 cereals andd flour nuts 6 0.013-0.135 0.034 breadd 2 0.025-0.030 0.020 cocoa 1 0.029 vegetables 44 0.01-0.8 0.047 a Micrograms of Se/gram of dry weight. *Not counting the data for Boletus edulis [17.34 pg of Se/g, similar to the results of Piepponen et al. (1983)], which would distort the mean.
Table IV. Contribution of Food of Vegetable Origin to Mean Selenium Intake bv the Galician PoDulation" ~~
sample fruit mushrooms cereals bread cocoa fresh green veg dry legumes potatoes total
mean, ppm 0.0019 0.051 0.022 0.017 0.028 0.0078 0.011 0.0044
~
g consumed/person Se intake, in per day pg/person per day Randoin OERGA Randoin OERGA 135 142.2 0.26 0.27 10 10 0.51 0.51 35 0.77 1.15 52.1 400 6.8 5.08 298.8 25 6.6 0.7 0.18 300 147.3 2.34 1.15 25 19 0.28 0.21 300 602 1.32 2.65 12.98 11.20
=Data obtained from fresh weights.
are eliminated by complexation with EDTA. Table I lists the results of the reproducibility study and Table I1 those of the accuracy study (the results of the accuracy experiments using Lactarius piperatus are also shown in Figure 1). Both reproducibility and accuracy are within accepted limits (5% relative error, 95-105% accuracy), and other experiments carried out in our laboratory have shown the complex formed, 4,5-benzopiazselenole, to be stable for at least 2 h. Table I11 lists the means and ranges of the selenium contents of the various groups of food studied. Of the 60 samples analyzed, the mushrooms had by far the highest selenium levels, followed by cocoa. The lowest selenium
J. Agric. Food Chem. 1988, 36, 295-299
contents were those of fruit. On the basis of the data of Table 111, together with similar data obtained for vegetables by Herrero-Latorre et al. (1986), the contributions of these kinds of food to the mean intake of selenium in Galicia/person per day have been estimated by multiplying the mean Se content of each group by the mean consumption of that kind of food in Galicia/person per day (OERGA, 1980). The results are shown in Table IV togeter with those of similar calculations carried out using the mean consumptions recommended in the balanced diet tables of Randoin et al. (1969). It may be noted that in spite of their high Se levels, mushrooms make a relatively modest contribution to total Se intake because their consumption is not widespread in Galicia. The total Se intake in food of vegetable origin, 11.20 hg/person per day, is very similar to the figure of 13.01 yg/person per day obtained using the tables of Randoin et al. (1969). The fact that both figures are considerably lower than those for other parts of the world [in Canada, for exaniple, cereals alone provide 62-112 pg of Se/person per day (Thompson et al., 1975)] may be attributed to the low selenium content of soils. Registry No. Se, 7782-49-2.
LITERATURE CITED Analytical Methods Committee, Royal Society of Chemistry, Analyst 1979, 104, 118.
Hartley, W. J. N.Z. J. Agric. 1959, 100, 259.
295
Herrero-Latorre,C.;-Mejuto-MarG,M. C.; Boll&-Rod&uez, M. H.; Bermejo-Martinez,F. Andes Bromutologia 1987,39(1),133. Ihnat, M. J. Assoc. Off. Anal. Chem. 1976, 59, 911. Instituto Nacional de Estadistica, Ministerio de Economia, Observatorio Estadistico Regional de Galicia (OERGA) La Alimentacidn en Galicia, Santiago de Compostela, 1980. IUPAC Pure Appl. Chem. 1984,56,645. Kubota, J.; Allaway, W. H.; Carter, D. L.; Cary, E. E.; Lazar, V. A. J. Agric. Food Chem. 1967,15,448. Levander, 0. A. J. Am. Diet. Assoc. 1975, 66, 338. Levander, 0. A. R o c . Symp. Selenium-Tellurium Enuiron. 1976, 26.
Mondragon, M. C.; Jaffe, W. G. Arch. Latinoam. Nutr. 1976,26, 341.
Morris, W. C.; Levander, 0. A. J. Nutr. 1970, 100, 1383. National Research Council, Agricultural Board, Committee on Animal Nutrition, Subcommitteeon Selenium (J.E. Oldfield, Chairman) Selenium in Nutrition; National Academy of Sciences: Washington, DC, 1971. Olson, 0. E.; Palmer, I. S. J. Food Sci. 1984, 49, 446. Pipponen, S.; Liukkonen-Lilja,E.; Kuusi, T. 2. Lebensm. Unter. Forsch. 1983, 177, 257.
Pipponen, S. Kem. Kemi. 1984,11, 180. Randoin; Legalli; Duchene In Rations Equilibrees, 8th ed.; Foucher, E. Ed.; Hygiene: Paris, 1969; Vol. I, p 35. Rosenfield, I.; Beath, 0. A. Selenium, Geobotany,Biochemistry, Toxicity and Nutrition; Academic: New York, 1964. Schwarz, K.; Foltz, C. M. J. Am. Chem. SOC.1957, 79, 3292. Thompson, J. N.; Erdoby, P.; Smith, C. J.Nutr. 1975,105, 274. Received for review July 10, 1986. Revised manuscript received February 3, 1987. Accepted October 15, 1987.
Green Leaf Headspace Volatiles from Nicotiana tabacum Lines of Different Trichome Morphology Roger A. Andemen,* Thomas R. Hamilton-Kemp, John H. Loughrin, Charles G. Hughes, David F. Hildebrand, and Tommy G. Sutton Volatile compounds from fully expanded whole leaves of four greenhouse-grown vegetative (green) tobacco lines were isolated by entrainment in purified air followed by Tenax trapping, characterized by capillary GC and GC-MS, and quantified by capillary GC. Identified headspace compounds, which comprised about 50% of the number of GC peaks and weight of the total estimated volatiles, were (E)-@-ocimene, (2)-3-hexenyl acetate, (2)-3-hexen-l-o1, linalool, P-caryophyllene, (E)-P-farnesene, solanone, methyl salicylate, nicotine, and neophytadiene. Yields of total estimated volatile compounds among the replicated leaf samples of the tobacco lines ranged from 6.6 to 38.5 ng/g wet weight. The means of amounts of volatile components that were believed to be emitted as a result of wound-induced lipoxygenase-hydroperoxide lyase activities were higher for analyses performed with 5-L entrainment flasks than with 12-L flasks. The leaf surface emissions for individual volatile compounds are discussed in terms of genetic differences including leaf trichome morphologies among Tobacco Introduction (TI) 1112, TI 1406, and TI 1068 and KY 14 burley. The results obtained do not support the view that the majority of headspace volatiles are emitted from exudate-secreting glandular heads of leaf trichomes. However, there were significant differences in yields of some compounds among genetic lines of tobacco. There have been relatively few reports on the occurrence, quantities, and biogenesis of volatile compounds in green vegetative tobacco as compared with cured tobacco. It is recognized, however, that knowledge about the composition of volatiles in tobacco prior to harvest could be USDA-ARS, MSA (R.A.A., T.G.S.), Departments of Agronomy (R.A.A., D.F.H., T.G.S.) and Horticulture (T.R.H.-K., J.H.L.), and Tobacco and Health Research Institute (C.G.H.), University of Kentucky, Lexington, Kentucky 40546. 002 1-856 118811436-0295$01.5010
useful in the understanding of the biogenesis of compounds in postharvest tobacco that affect the quality of leaf and may play a role in plant-pest interactions. The Occurrence of a t least 25 compounds in steam distillates and five compounds in the headspace of burley tobacco stalk and their quantitative analyses were recently reported (Andersen et al., 1986). Burton and Kasperbauer (1985) showed that the composition of some volatile components in green burley leaf at harvest changed during air-curing. Several reports have appeared on volatiles in terms of the effects of cultural practices on cured tobacco (Weeks and Seltmann, 1986),composition in flue-cured tobacco (Lloyd 0 1988 American Chemical Society
